0021-972X/91/7306-1377$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1991 by The Endocrine Society
Vol. 73, No. 6 Printed in U.S.A.
INSULIN-LIKE GROWTH FACTOR (IGF) BINDING PROTEIN-3 IN PREGNANCY SERUM BINDS NATIVE IGF-I BUT NOT IODO-IGF-L Anne-Maria Suikkari and Robert C. Baxter Department of Endocrinology, Royal Prince Alfred Hospital, Camperdown, NSW 2050, Australia. ABSTRACT. Although serum immunoreactive insulin-like growth factor binding protein-3 (IGFBP-3) increases during pregnancy, radioligand binding methods such as ligand blotting with iodinated IGFs fail to detect the protein in pregnancy serum. Since IGFBP-3 must bind IGF-I or IGF-II to form a complex with the acid-labile subunit (a-subunit), we have used ternary complex formation from [125Ijocsubunit as a measure of IGF binding to serum IGFBP-3. High-pressure liquid chromatography fractions containing IGFBP-3 from pregnancy serum did not bind [125I]IGF-I, although the equivalent fractions from nonpregnancy serum showed dose-dependent binding. In contrast, IGFBP-3 fractions from nonpregnancy and pregnancy sera both bound [125I]a-subunit in the presence of either exogenous IGF-I or endogenous serum IGFs, implying that non-iodinated IGFs could bind to the IGFBP-3. Substitution of nonradioactive iodo-IGF-I for native IGF-I in the complex formation assay confirmed that the pregnancy-induced alteration in IGFBP-3, probably resulting from proteolysis, prevents it from binding iodo-IGF-I while having little effect on its bindiog of the native peptide. This provides an explanation for the failure to detect IGFBP-3 in pregnancy by radioligand binding methods, and raises the question of the significance of proteolysis of IGFBP-3. pregnant and nonpregnant subjects were used in this study. Identical results were obtained from repeated experiments using individual serum samples. Fractionation of serum Serum samples (200 |0.L) were fractionated on a Superose-12 10/30 gel permeation column (Pharmacia, Uppsala, Sweden) at neutral pH, as previously described (5). The three 0.5 mL fractions eluting from 10.5-12 mL, which contain the 150K complex (5), were pooled and concentrated six-fold on Centricon™ 10 microconcentrators (Amicon, Danvers, MA). Each concentrated sample (240 |J.L) was adjusted to 0.2 M acetic acid, 0.1 M trimethylamine, pH 2.6 by addition of 60 |iL of 5fold concentrated buffer and after 5 min, a 200 (J.L aliquot was fractionated on a Protein Pak 125 column (Waters, Milford, MA) ainning at 1.7 mL/min in the same pH 2.6 buffer. Three fractions of 1.3 min (2.2 mL) were collected, one eluting from 3.9 - 5.2 min, containing IGFBPs, and two eluting from 5.2 - 7.8 min, containing IGFs and lower molecular weight substances (14). Each fraction was neutralized to pH 6.5 immediately with 1 M Tris base. Aliquots of the IGFBP fraction were used for the [125I]IGF-I binding and complex formation assays as indicated below. As controls, aliquots of the acidified serum concentrates, identical to those fractionated on HPLC, were diluted to an equivalent extent, but not fractionated. Tracer IGF binding assay Measurement of [125IJIGF-I binding activity were performed as described (12). Six concentrations (2.5 - 50 |a.L, 1:5 diluted) of the HPLC fraction containing IGFBP-3 were incubated with [ 125 I]IGF-I or -II (-10,000 cpm) for 2 h at 22 C. The IGFIGFBP-3 complexes were immunoprecipitated by 0.5 (0.L of specific IGFBP-3 antibody (R7). After incubation for 4 h at 22 C, 2 (iL of second antibody, goat antirabbit antiserum, was added, and after a further 30 min at 22 C, 1 ml 60 g/L cold PEG solution in 9 g/L NaCl was added. After centrifugation, radioactivity in the pellets was counted on a gamma counter. Ternary complex formation assay The ternary complex formation assay measured the ability of IGF-IGFBP-3 to combine with added tracer a-subunit (12). Serial dilutions of the HPLC fraction containing IGFBP-3 (2.5 100 (iL) were incubated with | 12S I|a-subunit (-10,000 cpm). The incubation and immunoprecipitation was carried out as above except that 1 ^.L of IGFBP-3 antibody and 4 |j.L of second antibody were added.
The majority of insulin-like growth factors (IGF-I and IGFII) circulate as a high molecular weight ternary complex (1-3). In addition to IGFs, the 150K complex contains an acid-stable binding protein, IGFBP-3 and an acid-labile non-lGF-binding asubunit (4). IGFBP-3 binds the a-subunit only when its IGFbinding site is occupied. The ternary complex is readily dissociated by acidification which destroys the a-subunit but leaves IGFBP-3 intact (5). In pregnancy serum, the immunoreactive IGFBP-3 concentration is increased compared to nonpregnancy serum (6). In contrast, IGFBP-3 appears to be reduced or absent in pregnancy seaim when measured by ligand blotting (7-9). This is thought to be caused by an increased protease activity in pregnancy serum (9, 10). However, the molecular weight distribution of IGFBP-3 is identical in pregnancy and nonpregnancy sera, with approximately 90% in the 150K complex (11). We have recently shown that, in serum samples which have been transiently acidified to destroy endogenous a-subunit, the re-binding of exogenous a-subunit to form the ternary complex appears normal, implying normal binding of IGFs to IGFBP-3 (11). In this study we have used the ternary complex formation assay as a measure of IGF binding to IGFBP-3, without using radioiodinated IGFs. Our results indicate that, whereas IGFBP-3 in nonpregnancy serum binds IGF-I and iodo-IGF-I similarly, IGFBP-3 in pregnancy serum is unable to bind iodo-IGF-I. MATERIALS AND METHODS Reagents IGF-I, IGF-II, and the a-subunit were purified as described (12). IGF-I, IGF-II, and a-subunit were iodinated using chloramine T (12, 13). Using the same method, nonradioactive iodo-IGF-I was prepared by reacting IGF-I (5 |ag) with nonradioactive sodium iodide (1 |ig). Subjects Blood samples were obtained from apparently healthy pregnant women attending the outpatient clinic in King George V Hospital, Camperdown, NSW between 28-40 weeks of gestation. Nonpregnancy serum samples were obtained from apparently healthy adult volunteers. Aliquots of serum were stored at -20 C until used. Pooled serum samples from five
1377
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 29 August 2015. at 03:05 For personal use only. No other uses without permission. . All rights reserved.
1378
RAPID COMMUNICATIONS
JCE & M • 1991 Vol73«No6
RESULTS After depletion of endogenous IGFs by HPLC at pH 2.6, the ability of IGFBP-3 in pregnancy serum to bind [125I]IGF-I was completely lost, while IGFBP-3 from nonpregnancy serum retained binding activity (Fig. la). Identical results were obtained when [125T]TGF_II w a s u s e c j a s t n e radioligand, or when unlabeled oc-subunit (50 ng/tube) was added to the binding reactions (not shown). In contrast, there was only a small difference in the ability of the same fractions containing IGFBP-3 to bind [125I]cc-subunit in the presence of exogenous IGF-I (Fig. lb). This implies that the IGFBP-3 from both pregnancy and nonpregnancy sera retained substantial IGF-binding activity, an observation apparently inconsistent with the first result.
40
a Nonpregnancy
b
10 100 assayed
Pregnancy
10 100 assayed
FIG. 2. Complex formation by nonpregnancy (a) and pregnancy (b) serum IGFBP-3 without exogenous IGFs. Serum samples were treated as described in the FIG. 1 legend. Another aliquot of the concentrated, acidified sample was incubated with HPLC buffer without fractionation. Serial dilutions of the acidified nonfractionated serum (O), the fractions containing IGFBP-3 alone ( • ) , and the mixed fractions containing IGFBP-3 and IGFs ( • ) were incubated with l 125 ljoc-subunit without exogenous IGFs as described in MATERALS AND METHODS. The experiment was repeated five times on either pooled or individual samples.
10 100 assayed
10 100 assayed
FIG. l. (a) [125I]IGF-I binding to IGFBP-3 from pooled (n = 5) nonpregnancy (D) and pregnancy ( • ) s e r a . Serum samples were fractionated on a Superose-12 column at neutral pH. Fractions corresponding to the 150K peaks were concentrated and subjected to HPLC to remove the endogenous IGFs. Increasing volumes (2.5 - 50 jiL) of 1:5 diluted fraction containing IGFBPs were incubated with [ 125I1IGF-I as described in MATERALS AND METHODS. The experiment was repeated five times on either pooled or individual samples, (b) Complex formation by IGFBP-3 from similarly processed 150K peaks from nonpregnancy (D) and pregnancy ( • ) sera. Serial dilutions of fractions containing IGFBP-3 were incubated with [125I]cc-subunit in the presence of a constant amount of IGF-I (10 ng/tube). The experiment was repeated four times on either pooled or individual samples. To determine whether endogenous IGFs from pregnancy and nonpregnancy sera could participate in ternary complex formation as effectively as exogenous IGF-I, HPLC fractions containing IGFs and lower molecular weight substances were recombined with the fraction containing IGFBP-3. Fig. 2a shows [ 125 I]asubunit binding to increasing concentrations of the IGFBP-3 fraction from nonpregnancy serum, added alone or in combination with the lower molecular weight fractions containing IGFs. Ternary complex formation was minimal in the absence of added IGFs. However, when the IGF fractions were added to the IGFBP-3 fraction, [125I]a-subunit binding was fully restored to the level seen in a similar sample that had been acidified but not fractionated. Similar results for pregnancy serum are shown in Fig. 2b. While [125I]oc-subunit binding could be restored by the addition of endogenous IGFs, there was a consistent small difference between the binding to the reconstituted sample and the sample that had been acidified but not fractionated.
We then investigated whether the discrepancy between binary and ternary complex formation in pregnancy serum might be due to the iodination of IGF-I used as a ligand in the binary complex formation assay. Since ternary complex formation involves IGF binding to IGFBP-3, the ternary complex formation assay can be used as a measure of IGF binding without the necessity of using iodinated IGFs. The IGF-I dose-response relationship for asubunit binding to the IGFBP-3 from nonpregnancy serum was similar for both native IGF-I and iodo-IGF-I (Fig. 3a). However, the pregnancy serum IGFBP-3, while forming the ternary complex with native IGF-I, was unable to form the complex with iodo-IGF-I in five experiments (Fig. 3b). 60
a Nonpregnancy
b
Pregnancy
P°
.1 1 ng IGF-I
10
FIG. 3. IGF-I dose responses for [ 125 I]a-subunit binding to pooled (n = 5) nonpregnancy (a) and pregnancy (b) samples. A constant amount (50 (i.L) of the neutralized HPLC fraction containing IGFBP-3 was incubated with [ 125 I]a-subunit and graded concentrations (0.05 - 10 ng) of IGF-I (D) o r nonradioactive iodo-IGF-I ( • ) as described in MATERALS AND METHODS. The experiment was repeated five times on either pooled or individual samples.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 29 August 2015. at 03:05 For personal use only. No other uses without permission. . All rights reserved.
RAPID COMMUNICATIONS DISCUSSION It has previously been demonstrated that cc-subunit binds only to IGF-IGFBP-3 binary complex, but not to IGFBP-3 alone. (4). The measurement of ternary complex formation thus provides a direct estimate of the oc-subunit binding activity of IGFBP-3, and an indirect estimate of the IGF-I binding activity. Several independent studies have demonstrated that IGFBP-3 from pregnancy serum, when stripped of its endogenous ligands, has a greatly reduced ability to rebind radioiodinated IGFs (7-9, 15 16). These reports are hard to reconcile with our recent observation that pregnancy serum, acidified to destroy its endogenous a-subunit, is able to bind tracer a-subunit in the presence of its endogenous IGFs, implying normal IGF binding by pregnancy IGFBP-3 (11). There are several possible explanations for the discrepancy between direct IGF binding studies (i.e. those using radioiodinated ligands) and studies in which IGF binding is inferred from the formation of the ternary complex. First, it may be necessary to physically separate the IGFs from IGFBP-3 in pregnancy serum (e.g. by chromatography or electrophoresis) before the loss of IGF binding becomes apparent. Second, the presence of a-subunit might be necessary for IGFs to rebind to the IGF-depleted pregnancy serum. Third, experiments using iodinated IGFs may not accurately reflect the binding activity of native IGFs. In the present study we have refuted the first possibility by showing that, even after the IGFs are physically removed from the IGFBP-3 by chromatography at low pH, the ternary complex reforms in both pregnancy and nonpregnancy sera, in the presence of either exogenous IGF-I or endogenous IGFs. A slight loss of activity after this procedure was, however, consistently seen in pregnancy serum. We also found that addition of pure a-subunit had no effect on tracer IGF-I binding to IGFBP-3 from either pregnancy or nonpregnancy sera. However, a comparison of native and iodo-IGF-I in the ternary complex formation assay revealed that, whereas IGFBP-3 from nonpregnancy serum reacted with the two IGFs similarly, iodoIGF-I was unable to participate in ternary complex formation with IGFBP-3 from pregnancy serum. This indicates that IGFBP-3 in pregnancy serum cannot bind iodo-IGF-I, and provides a simple explanation for the discrepancy between binding studies using iodinated IGFs and those using the native peptides. It is likely that a pregnancy protease acts on IGFBP-3 to render it unable to bind iodinated IGFs (8-10). Although the ability of proteolyzed IGFBP-3 to bind native IGFs may also be slightly reduced, our studies indicate that IGFBP-3 from pregnancy serum retains most of its IGF binding activity. Thus the physiological significance of this limited proteolysis is unclear. While it can be argued that increased IGF availability might have advantages in pregnancy, it remains to be demonstrated whether the minimal reduction in the IGF binding activity of pregnancy IGFBP-3, which appears exacerbated in experiments using iodinated IGFs has any functional conseciuence. ACKNOWLEDGEMENTS
1379
REFERENCES 1. Furlanetto RW. The somatomedin C binding protein: evidence for a heterologous subunit structure. J Clin Endocrinol Metab. 1980;51:12-19. 2. Hintz RL, Liu F, Rosenfeld RG, Kemp SF. Plasma somatomedin-binding proteins in hypopituitarism: changes during growth hormone therapy. J Clin Endocrinol Metab. 1981;53:100-104. 3. Daughaday WH, Ward AP, Goldberg AC, Trivedi B, Kapadia M. Characterization of somatomedin binding in human serum by ultracentrifugation and gel filtration. J Clin Endocrinol Metab. 1982:55:916-921. 4. Baxter RC, Martin JL. Structure of the Mr 140,000 growth hormone-dependent insulin-like growth factor binding protein complex: Determination by reconstiiution and affinity labeling. Proc Natl Acad Sci USA. 1989:86:6898-6902. 5. Baxter RC. Characterization of the acid-labile subunit of the growth hormone-dependent insulin-like growth factor binding protein complex. J Clin Endocrinol Metab. 1988;67:265-272. 6. Baxter RC, Martin JL. Radioimmunoassay of growth hormone-dependent insulinlike growth factor binding protein in human plasma. J Clin Invest. 1986;78:1504-1512. 7. Gargosky SE, Moyse KJ, Walton PE, Owens JA, Wallace JC, Robinson JS, Owens PC. Circulating levels of insulinlike growth factors increase and molecular forms of their binding proteins change with human pregnancy. Biochem Biophys Res Commun. 1990:170:1157-1163. 8. Giudice LC, Farrell EM, Pham H, Lamson G, Rosenfeld RG. Insulin-like growth factor binding proteins in maternal serum throughout gestation and in the puerperium: effects of a pregnancy-associated protease activity. J Clin Endocrinol Metab. 1990:71:806-816. 9. Hossenlopp P, Segovia B, Lassarre C, Roghani M, Bred on M, Binoux M. Evidence of enzymatic degradation of insulinlike growth factor-binding proteins in the 150K complex during pregnancy. J Clin Endocrinol Metab. 1990;71:797805. 10. Lamson G, Giudice LC, Rosenfeld RG. A simple assay for proteolysis of IGFBP-3. J Clin Endocrinol Metab. 1991:72:1391-1393. 11. Suikkari A-M, Baxter RC. Insulin-like growth factor binding protein-3 (IGFBP-3) is functionally normal in pregnancy serum. J Clin Endocrinol Metab. 1991 ;in press. 12. Baxter RC, Martin JL, Beniac VA. High molecular weight insulin-like growth factor binding protein complex. Purification and properties of the acid-labile subunit from human serum. J Biol Chem. 1989:264:11843-11848. 13. Baxter RC. Circulating levels and molecular distribution of the acid-labile (a) subunit of the high molecular weight insulin-like growth factor-binding protein complex. J Clin Endocrinol Metab. 1990:70:1347-1353. 14. Scott CD, Martin JL, Baxter RC. Production of insulin-like growth factor I and its binding protein by adult rat hepatocytes in primary culture. Endocrinology. 1985:116:1094-1101. 15. Davenport ML, Clemmons DR, Miles MV, CamachoHubner C, D'Ercole AJ, Underwood LE. Regulation of serum insulin-like growth factor-I (IGF-I) and IGF binding proteins during rat pregnancy. Endocrinology. 1990:127:1278-1286. 16. Fielder PJ, Thordarson G, Talamantes F, Rosenfeld RG. Characterization of insulin-like growth factor binding proteins (IGFBPs) during gestation in mice: effects of hypophysectomy and an IGFBP-specific serum protease activity. Endocrinology. 1990; 127:2270-2280.
This work was supported by grants from the Finnish Cultural Foundation, Jalmari and Rauha Ahokas Foundation, Hoechst Australia, and The National Health and Medical Research Council, Australia.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 29 August 2015. at 03:05 For personal use only. No other uses without permission. . All rights reserved.
Vol. 73, No. 6 Printed in U.S.A.
INSULIN-LIKE GROWTH FACTOR (IGF) BINDING PROTEIN-3 IN PREGNANCY SERUM BINDS NATIVE IGF-I BUT NOT IODO-IGF-L Anne-Maria Suikkari and Robert C. Baxter Department of Endocrinology, Royal Prince Alfred Hospital, Camperdown, NSW 2050, Australia. ABSTRACT. Although serum immunoreactive insulin-like growth factor binding protein-3 (IGFBP-3) increases during pregnancy, radioligand binding methods such as ligand blotting with iodinated IGFs fail to detect the protein in pregnancy serum. Since IGFBP-3 must bind IGF-I or IGF-II to form a complex with the acid-labile subunit (a-subunit), we have used ternary complex formation from [125Ijocsubunit as a measure of IGF binding to serum IGFBP-3. High-pressure liquid chromatography fractions containing IGFBP-3 from pregnancy serum did not bind [125I]IGF-I, although the equivalent fractions from nonpregnancy serum showed dose-dependent binding. In contrast, IGFBP-3 fractions from nonpregnancy and pregnancy sera both bound [125I]a-subunit in the presence of either exogenous IGF-I or endogenous serum IGFs, implying that non-iodinated IGFs could bind to the IGFBP-3. Substitution of nonradioactive iodo-IGF-I for native IGF-I in the complex formation assay confirmed that the pregnancy-induced alteration in IGFBP-3, probably resulting from proteolysis, prevents it from binding iodo-IGF-I while having little effect on its bindiog of the native peptide. This provides an explanation for the failure to detect IGFBP-3 in pregnancy by radioligand binding methods, and raises the question of the significance of proteolysis of IGFBP-3. pregnant and nonpregnant subjects were used in this study. Identical results were obtained from repeated experiments using individual serum samples. Fractionation of serum Serum samples (200 |0.L) were fractionated on a Superose-12 10/30 gel permeation column (Pharmacia, Uppsala, Sweden) at neutral pH, as previously described (5). The three 0.5 mL fractions eluting from 10.5-12 mL, which contain the 150K complex (5), were pooled and concentrated six-fold on Centricon™ 10 microconcentrators (Amicon, Danvers, MA). Each concentrated sample (240 |J.L) was adjusted to 0.2 M acetic acid, 0.1 M trimethylamine, pH 2.6 by addition of 60 |iL of 5fold concentrated buffer and after 5 min, a 200 (J.L aliquot was fractionated on a Protein Pak 125 column (Waters, Milford, MA) ainning at 1.7 mL/min in the same pH 2.6 buffer. Three fractions of 1.3 min (2.2 mL) were collected, one eluting from 3.9 - 5.2 min, containing IGFBPs, and two eluting from 5.2 - 7.8 min, containing IGFs and lower molecular weight substances (14). Each fraction was neutralized to pH 6.5 immediately with 1 M Tris base. Aliquots of the IGFBP fraction were used for the [125I]IGF-I binding and complex formation assays as indicated below. As controls, aliquots of the acidified serum concentrates, identical to those fractionated on HPLC, were diluted to an equivalent extent, but not fractionated. Tracer IGF binding assay Measurement of [125IJIGF-I binding activity were performed as described (12). Six concentrations (2.5 - 50 |a.L, 1:5 diluted) of the HPLC fraction containing IGFBP-3 were incubated with [ 125 I]IGF-I or -II (-10,000 cpm) for 2 h at 22 C. The IGFIGFBP-3 complexes were immunoprecipitated by 0.5 (0.L of specific IGFBP-3 antibody (R7). After incubation for 4 h at 22 C, 2 (iL of second antibody, goat antirabbit antiserum, was added, and after a further 30 min at 22 C, 1 ml 60 g/L cold PEG solution in 9 g/L NaCl was added. After centrifugation, radioactivity in the pellets was counted on a gamma counter. Ternary complex formation assay The ternary complex formation assay measured the ability of IGF-IGFBP-3 to combine with added tracer a-subunit (12). Serial dilutions of the HPLC fraction containing IGFBP-3 (2.5 100 (iL) were incubated with | 12S I|a-subunit (-10,000 cpm). The incubation and immunoprecipitation was carried out as above except that 1 ^.L of IGFBP-3 antibody and 4 |j.L of second antibody were added.
The majority of insulin-like growth factors (IGF-I and IGFII) circulate as a high molecular weight ternary complex (1-3). In addition to IGFs, the 150K complex contains an acid-stable binding protein, IGFBP-3 and an acid-labile non-lGF-binding asubunit (4). IGFBP-3 binds the a-subunit only when its IGFbinding site is occupied. The ternary complex is readily dissociated by acidification which destroys the a-subunit but leaves IGFBP-3 intact (5). In pregnancy serum, the immunoreactive IGFBP-3 concentration is increased compared to nonpregnancy serum (6). In contrast, IGFBP-3 appears to be reduced or absent in pregnancy seaim when measured by ligand blotting (7-9). This is thought to be caused by an increased protease activity in pregnancy serum (9, 10). However, the molecular weight distribution of IGFBP-3 is identical in pregnancy and nonpregnancy sera, with approximately 90% in the 150K complex (11). We have recently shown that, in serum samples which have been transiently acidified to destroy endogenous a-subunit, the re-binding of exogenous a-subunit to form the ternary complex appears normal, implying normal binding of IGFs to IGFBP-3 (11). In this study we have used the ternary complex formation assay as a measure of IGF binding to IGFBP-3, without using radioiodinated IGFs. Our results indicate that, whereas IGFBP-3 in nonpregnancy serum binds IGF-I and iodo-IGF-I similarly, IGFBP-3 in pregnancy serum is unable to bind iodo-IGF-I. MATERIALS AND METHODS Reagents IGF-I, IGF-II, and the a-subunit were purified as described (12). IGF-I, IGF-II, and a-subunit were iodinated using chloramine T (12, 13). Using the same method, nonradioactive iodo-IGF-I was prepared by reacting IGF-I (5 |ag) with nonradioactive sodium iodide (1 |ig). Subjects Blood samples were obtained from apparently healthy pregnant women attending the outpatient clinic in King George V Hospital, Camperdown, NSW between 28-40 weeks of gestation. Nonpregnancy serum samples were obtained from apparently healthy adult volunteers. Aliquots of serum were stored at -20 C until used. Pooled serum samples from five
1377
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 29 August 2015. at 03:05 For personal use only. No other uses without permission. . All rights reserved.
1378
RAPID COMMUNICATIONS
JCE & M • 1991 Vol73«No6
RESULTS After depletion of endogenous IGFs by HPLC at pH 2.6, the ability of IGFBP-3 in pregnancy serum to bind [125I]IGF-I was completely lost, while IGFBP-3 from nonpregnancy serum retained binding activity (Fig. la). Identical results were obtained when [125T]TGF_II w a s u s e c j a s t n e radioligand, or when unlabeled oc-subunit (50 ng/tube) was added to the binding reactions (not shown). In contrast, there was only a small difference in the ability of the same fractions containing IGFBP-3 to bind [125I]cc-subunit in the presence of exogenous IGF-I (Fig. lb). This implies that the IGFBP-3 from both pregnancy and nonpregnancy sera retained substantial IGF-binding activity, an observation apparently inconsistent with the first result.
40
a Nonpregnancy
b
10 100 assayed
Pregnancy
10 100 assayed
FIG. 2. Complex formation by nonpregnancy (a) and pregnancy (b) serum IGFBP-3 without exogenous IGFs. Serum samples were treated as described in the FIG. 1 legend. Another aliquot of the concentrated, acidified sample was incubated with HPLC buffer without fractionation. Serial dilutions of the acidified nonfractionated serum (O), the fractions containing IGFBP-3 alone ( • ) , and the mixed fractions containing IGFBP-3 and IGFs ( • ) were incubated with l 125 ljoc-subunit without exogenous IGFs as described in MATERALS AND METHODS. The experiment was repeated five times on either pooled or individual samples.
10 100 assayed
10 100 assayed
FIG. l. (a) [125I]IGF-I binding to IGFBP-3 from pooled (n = 5) nonpregnancy (D) and pregnancy ( • ) s e r a . Serum samples were fractionated on a Superose-12 column at neutral pH. Fractions corresponding to the 150K peaks were concentrated and subjected to HPLC to remove the endogenous IGFs. Increasing volumes (2.5 - 50 jiL) of 1:5 diluted fraction containing IGFBPs were incubated with [ 125I1IGF-I as described in MATERALS AND METHODS. The experiment was repeated five times on either pooled or individual samples, (b) Complex formation by IGFBP-3 from similarly processed 150K peaks from nonpregnancy (D) and pregnancy ( • ) sera. Serial dilutions of fractions containing IGFBP-3 were incubated with [125I]cc-subunit in the presence of a constant amount of IGF-I (10 ng/tube). The experiment was repeated four times on either pooled or individual samples. To determine whether endogenous IGFs from pregnancy and nonpregnancy sera could participate in ternary complex formation as effectively as exogenous IGF-I, HPLC fractions containing IGFs and lower molecular weight substances were recombined with the fraction containing IGFBP-3. Fig. 2a shows [ 125 I]asubunit binding to increasing concentrations of the IGFBP-3 fraction from nonpregnancy serum, added alone or in combination with the lower molecular weight fractions containing IGFs. Ternary complex formation was minimal in the absence of added IGFs. However, when the IGF fractions were added to the IGFBP-3 fraction, [125I]a-subunit binding was fully restored to the level seen in a similar sample that had been acidified but not fractionated. Similar results for pregnancy serum are shown in Fig. 2b. While [125I]oc-subunit binding could be restored by the addition of endogenous IGFs, there was a consistent small difference between the binding to the reconstituted sample and the sample that had been acidified but not fractionated.
We then investigated whether the discrepancy between binary and ternary complex formation in pregnancy serum might be due to the iodination of IGF-I used as a ligand in the binary complex formation assay. Since ternary complex formation involves IGF binding to IGFBP-3, the ternary complex formation assay can be used as a measure of IGF binding without the necessity of using iodinated IGFs. The IGF-I dose-response relationship for asubunit binding to the IGFBP-3 from nonpregnancy serum was similar for both native IGF-I and iodo-IGF-I (Fig. 3a). However, the pregnancy serum IGFBP-3, while forming the ternary complex with native IGF-I, was unable to form the complex with iodo-IGF-I in five experiments (Fig. 3b). 60
a Nonpregnancy
b
Pregnancy
P°
.1 1 ng IGF-I
10
FIG. 3. IGF-I dose responses for [ 125 I]a-subunit binding to pooled (n = 5) nonpregnancy (a) and pregnancy (b) samples. A constant amount (50 (i.L) of the neutralized HPLC fraction containing IGFBP-3 was incubated with [ 125 I]a-subunit and graded concentrations (0.05 - 10 ng) of IGF-I (D) o r nonradioactive iodo-IGF-I ( • ) as described in MATERALS AND METHODS. The experiment was repeated five times on either pooled or individual samples.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 29 August 2015. at 03:05 For personal use only. No other uses without permission. . All rights reserved.
RAPID COMMUNICATIONS DISCUSSION It has previously been demonstrated that cc-subunit binds only to IGF-IGFBP-3 binary complex, but not to IGFBP-3 alone. (4). The measurement of ternary complex formation thus provides a direct estimate of the oc-subunit binding activity of IGFBP-3, and an indirect estimate of the IGF-I binding activity. Several independent studies have demonstrated that IGFBP-3 from pregnancy serum, when stripped of its endogenous ligands, has a greatly reduced ability to rebind radioiodinated IGFs (7-9, 15 16). These reports are hard to reconcile with our recent observation that pregnancy serum, acidified to destroy its endogenous a-subunit, is able to bind tracer a-subunit in the presence of its endogenous IGFs, implying normal IGF binding by pregnancy IGFBP-3 (11). There are several possible explanations for the discrepancy between direct IGF binding studies (i.e. those using radioiodinated ligands) and studies in which IGF binding is inferred from the formation of the ternary complex. First, it may be necessary to physically separate the IGFs from IGFBP-3 in pregnancy serum (e.g. by chromatography or electrophoresis) before the loss of IGF binding becomes apparent. Second, the presence of a-subunit might be necessary for IGFs to rebind to the IGF-depleted pregnancy serum. Third, experiments using iodinated IGFs may not accurately reflect the binding activity of native IGFs. In the present study we have refuted the first possibility by showing that, even after the IGFs are physically removed from the IGFBP-3 by chromatography at low pH, the ternary complex reforms in both pregnancy and nonpregnancy sera, in the presence of either exogenous IGF-I or endogenous IGFs. A slight loss of activity after this procedure was, however, consistently seen in pregnancy serum. We also found that addition of pure a-subunit had no effect on tracer IGF-I binding to IGFBP-3 from either pregnancy or nonpregnancy sera. However, a comparison of native and iodo-IGF-I in the ternary complex formation assay revealed that, whereas IGFBP-3 from nonpregnancy serum reacted with the two IGFs similarly, iodoIGF-I was unable to participate in ternary complex formation with IGFBP-3 from pregnancy serum. This indicates that IGFBP-3 in pregnancy serum cannot bind iodo-IGF-I, and provides a simple explanation for the discrepancy between binding studies using iodinated IGFs and those using the native peptides. It is likely that a pregnancy protease acts on IGFBP-3 to render it unable to bind iodinated IGFs (8-10). Although the ability of proteolyzed IGFBP-3 to bind native IGFs may also be slightly reduced, our studies indicate that IGFBP-3 from pregnancy serum retains most of its IGF binding activity. Thus the physiological significance of this limited proteolysis is unclear. While it can be argued that increased IGF availability might have advantages in pregnancy, it remains to be demonstrated whether the minimal reduction in the IGF binding activity of pregnancy IGFBP-3, which appears exacerbated in experiments using iodinated IGFs has any functional conseciuence. ACKNOWLEDGEMENTS
1379
REFERENCES 1. Furlanetto RW. The somatomedin C binding protein: evidence for a heterologous subunit structure. J Clin Endocrinol Metab. 1980;51:12-19. 2. Hintz RL, Liu F, Rosenfeld RG, Kemp SF. Plasma somatomedin-binding proteins in hypopituitarism: changes during growth hormone therapy. J Clin Endocrinol Metab. 1981;53:100-104. 3. Daughaday WH, Ward AP, Goldberg AC, Trivedi B, Kapadia M. Characterization of somatomedin binding in human serum by ultracentrifugation and gel filtration. J Clin Endocrinol Metab. 1982:55:916-921. 4. Baxter RC, Martin JL. Structure of the Mr 140,000 growth hormone-dependent insulin-like growth factor binding protein complex: Determination by reconstiiution and affinity labeling. Proc Natl Acad Sci USA. 1989:86:6898-6902. 5. Baxter RC. Characterization of the acid-labile subunit of the growth hormone-dependent insulin-like growth factor binding protein complex. J Clin Endocrinol Metab. 1988;67:265-272. 6. Baxter RC, Martin JL. Radioimmunoassay of growth hormone-dependent insulinlike growth factor binding protein in human plasma. J Clin Invest. 1986;78:1504-1512. 7. Gargosky SE, Moyse KJ, Walton PE, Owens JA, Wallace JC, Robinson JS, Owens PC. Circulating levels of insulinlike growth factors increase and molecular forms of their binding proteins change with human pregnancy. Biochem Biophys Res Commun. 1990:170:1157-1163. 8. Giudice LC, Farrell EM, Pham H, Lamson G, Rosenfeld RG. Insulin-like growth factor binding proteins in maternal serum throughout gestation and in the puerperium: effects of a pregnancy-associated protease activity. J Clin Endocrinol Metab. 1990:71:806-816. 9. Hossenlopp P, Segovia B, Lassarre C, Roghani M, Bred on M, Binoux M. Evidence of enzymatic degradation of insulinlike growth factor-binding proteins in the 150K complex during pregnancy. J Clin Endocrinol Metab. 1990;71:797805. 10. Lamson G, Giudice LC, Rosenfeld RG. A simple assay for proteolysis of IGFBP-3. J Clin Endocrinol Metab. 1991:72:1391-1393. 11. Suikkari A-M, Baxter RC. Insulin-like growth factor binding protein-3 (IGFBP-3) is functionally normal in pregnancy serum. J Clin Endocrinol Metab. 1991 ;in press. 12. Baxter RC, Martin JL, Beniac VA. High molecular weight insulin-like growth factor binding protein complex. Purification and properties of the acid-labile subunit from human serum. J Biol Chem. 1989:264:11843-11848. 13. Baxter RC. Circulating levels and molecular distribution of the acid-labile (a) subunit of the high molecular weight insulin-like growth factor-binding protein complex. J Clin Endocrinol Metab. 1990:70:1347-1353. 14. Scott CD, Martin JL, Baxter RC. Production of insulin-like growth factor I and its binding protein by adult rat hepatocytes in primary culture. Endocrinology. 1985:116:1094-1101. 15. Davenport ML, Clemmons DR, Miles MV, CamachoHubner C, D'Ercole AJ, Underwood LE. Regulation of serum insulin-like growth factor-I (IGF-I) and IGF binding proteins during rat pregnancy. Endocrinology. 1990:127:1278-1286. 16. Fielder PJ, Thordarson G, Talamantes F, Rosenfeld RG. Characterization of insulin-like growth factor binding proteins (IGFBPs) during gestation in mice: effects of hypophysectomy and an IGFBP-specific serum protease activity. Endocrinology. 1990; 127:2270-2280.
This work was supported by grants from the Finnish Cultural Foundation, Jalmari and Rauha Ahokas Foundation, Hoechst Australia, and The National Health and Medical Research Council, Australia.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 29 August 2015. at 03:05 For personal use only. No other uses without permission. . All rights reserved.
